MOTIVATION LETTER The Engineering Biological and Biomedical Systems programme at NSF funds projects that control biological function through engineered systems. My independent research since 2025 has produced a tripartite pharmacological framework, the Conjunctive Consolidation Threshold model, that precisely controls reward-memory encoding in addiction. The CCT model reduces encoding probability from 0.855 to 0.122, an 85.8 percent reduction, with super-additivity of 12.8 percentage points across three drug classes. Five pre-registered hypotheses H1 through H5 were confirmed using ODE/RK45 numerical integration and Bayesian MCMC validation. This is not a drug discovery project. It is an engineering platform for preventing maladaptive learning. I have built three computational tools that operationalize this framework. IMPRINT screens addiction liability using the CCT equations. TOPOLOGIX applies persistent homology and bipartite simplicial complexes to drug-protein interaction networks, with a working MVP for hERG cardiotoxicity prediction. GATE evaluates safety of BCI neural-stimulation protocols under Apache 2.0 license. Together these form a closed-loop system: IMPRINT identifies risk, TOPOLOGIX maps molecular mechanisms, GATE ensures stimulation safety. The EBBS programme funds exactly this kind of biological function control through engineered platforms. My background combines pharmaceutical training with computational rigor. B.Pharm from University of Ibadan, CGPA 5.1 out of 7.0, German equivalent 1.9. Licensed pharmacist, Pharmaceutical Council of Nigeria. Current National Product Manager at Synthcare. Technical skills span Python, PyMC, MCMC, TDA with Ripser and Gudhi, NEURON and Brian2 for neural simulation, AlphaFold, RDKit, ADMET and QSAR, GROMACS, AutoDock, Nextflow and SLURM for HPC. I have endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard who provided my arXiv endorsement, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. A provisional patent on the CCT core architecture is filed for Q3 2026. I am a Nigerian independent researcher, 29 years old, based in Lagos. I am applying for MSc programs starting October 2026 at Medical University of Graz and University of Graz, Austria. This proposal requires a U.S. academic host as Principal Investigator. I have established relationships with potential hosts at Michigan, Harvard, Princeton, and NYU. The EBBS programme offers the funding mechanism to transition my independent work into a collaborative U.S.-based engineering project with direct application to the addiction crisis in Nigeria and globally. RESEARCH STATEMENT Project Title: An Engineered Closed-Loop Platform for Preventing Reward-Memory Encoding in Addiction Using the Conjunctive Consolidation Threshold Model The addiction crisis in Nigeria and across low- and middle-income countries lacks engineered solutions. Existing interventions target neurotransmitter systems individually, ignoring the conjunctive encoding mechanism that binds reward signals to memory consolidation. My CCT model demonstrates mathematically that three pharmacological interventions administered at specific temporal windows prevent this binding. The EBBS programme supports engineering of biological systems, not drug discovery. This project proposes to engineer a closed-loop platform that predicts, prevents, and monitors reward-memory encoding in real time. The CCT model specifies three interacting thresholds: a dopamine-dependent reward salience threshold, a glutamate-dependent consolidation threshold, and a norepinephrine-dependent arousal threshold. Encoding occurs only when all three thresholds are crossed simultaneously. The formal mathematical specification on OSF at DOI 10.17605/OSF.IO/EMY4U defines this as a system of coupled ordinary differential equations. Numerical integration using RK45 with adaptive step sizing shows that subthreshold dosing of a D1 antagonist, an NMDA partial agonist, and a beta-blocker reduces encoding probability from 0.855 to 0.122. Bayesian MCMC with PyMC confirms parameter identifiability across 10,000 posterior samples. The engineering challenge is threefold. First, IMPRINT must screen individual patients for their specific threshold values using pharmacokinetic and pharmacodynamic data. Second, TOPOLOGIX must verify that the three drug classes do not produce off-target topological disruptions in protein interaction networks. Persistent homology analysis of bipartite simplicial complexes from drug-protein docking simulations will quantify homology group persistence across dose ranges. Third, GATE must evaluate whether concurrent BCI neural stimulation alters the safety profile of the triple intervention. The GATE platform already simulates stimulation protocols and flags arrhythmia risk using hERG models. The proposed work during a 24-month funding period includes: completing the review article under consideration at Neuroscience and Biobehavioral Reviews; extending the Bayesian population dynamics model to include Nigerian pharmacokinetic data from published literature; validating TOPOLOGIX on the three drug classes against known cardiotoxicity databases; integrating IMPRINT, TOPOLOGIX, and GATE into a single pipeline with Supabase backend; and preparing a clinical trial architecture for a Phase I safety study in Lagos. The clinical trial architecture preprint on Zenodo at DOI 10.5281/zenodo.20492472 provides the regulatory framework. The broader impact addresses a specific need. Nigeria has no approved pharmacotherapy for methamphetamine or cannabis use disorders. The CCT platform is agnostic to the addictive substance because it targets the encoding mechanism common to all drugs of abuse. If successful, this platform can be deployed in low-resource settings where existing addiction treatments are unavailable or unaffordable. The EBBS programme explicitly funds projects that control biological function through engineering. This project does exactly that. PROJECT DESCRIPTION Specific Aim 1: Complete and publish the CCT mathematical framework and Bayesian validation. The foundational CCT paper on OSF at DOI 10.17605/OSF.IO/KG7B5 and the formal mathematical specification at DOI 10.17605/OSF.IO/EMY4U require peer review. The review article under consideration at Neuroscience and Biobehavioral Reviews will be revised based on reviewer feedback. The Bayesian population dynamics model on Zenodo at DOI 10.5281/zenodo.20492472 will be extended to include individual variability in metabolic enzyme activity using CYP2D6 and CYP3A4 genotype frequency data from the Nigerian population. Sensitivity analysis using Sobol indices will identify which parameters most affect encoding probability reduction. Deliverable: three published preprints and one accepted review article. Specific Aim 2: Validate TOPOLOGIX for the three CCT drug classes. TOPOLOGIX applies persistent homology to drug-protein interaction networks. Bipartite simplicial complexes are constructed from docking scores generated by AutoDock Vina for each drug against a panel of 50 cardiac ion channels and 50 central nervous system targets. Persistent homology barcodes are computed using Ripser and Gudhi. The null hypothesis is that no topological feature persists across dose ranges that would indicate off-target binding. The hERG cardiotoxicity MVP will be validated against the Zhan et al. 2024 benchmark dataset. Deliverable: a validated TOPOLOGIX module for the three drug classes with published barcode data. Specific Aim 3: Integrate GATE safety evaluation into the platform. GATE currently simulates BCI neural stimulation protocols and evaluates cardiac safety using a Hodgkin-Huxley model of ventricular myocytes with hERG current. This aim extends GATE to simulate concurrent administration of the three CCT drugs during stimulation. The NEURON simulation environment will model cortical and striatal neural populations under D1 antagonism, NMDA partial agonism, and beta-blockade. Safety endpoints include action potential duration prolongation, early afterdepolarizations, and seizure threshold reduction. Deliverable: GATE version 2.0 with integrated CCT pharmacology module. Specific Aim 4: Design and simulate a Phase I clinical trial protocol for Lagos. The clinical trial architecture on Zenodo specifies a three-arm, double-blind, placebo-controlled design with 36 healthy volunteers. This aim refines the protocol for Nigerian regulatory submission to the National Agency for Food and Drug Administration and Control. Pharmacokinetic simulations using a physiologically based model will determine dosing intervals that maintain subthreshold concentrations for all three drugs. The trial will measure encoding probability using a cue-reactivity paradigm with fMRI if available, or behavioral measures if not. Deliverable: a complete clinical trial protocol ready for IRB submission. TIMELINE Months 1-6: Revise and resubmit review article. Extend Bayesian model with Nigerian CYP data. Begin TOPOLOGIX validation on hERG benchmark. Months 7-12: Complete TOPOLOGIX validation. Begin GATE integration. Draft clinical trial protocol. Months 13-18: Complete GATE integration. Submit TOPOLOGIX results for preprint. Finalize clinical trial protocol. Months 19-24: Prepare integrated platform documentation. Submit Phase I protocol to NAFDAC. Write final report and submit for publication. BUDGET JUSTIFICATION Personnel: One graduate research assistant at 0.5 FTE for 24 months to assist with TOPOLOGIX validation and GATE integration. Computational costs: cloud HPC credits for Bayesian MCMC sampling and persistent homology calculations, estimated at 15,000 dollars. Software licenses: none required, all tools are open source. Publication costs: 3,000 dollars for open access fees. Travel: 5,000 dollars for one conference presentation and one collaborator meeting. Equipment: one workstation with GPU for neural simulations, 4,000 dollars. Total direct costs: 72,000 dollars. Indirect costs at the host institution standard rate. CHECKLIST - [ ] Identify a U.S. academic host institution and Principal Investigator willing to submit the proposal - [ ] Confirm host institution eligibility for NSF funding - [ ] Obtain letter of support from proposed PI (Berridge, Gershman, Daw, or Mattar) - [ ] Write and upload project narrative as described above - [ ] Prepare biographical sketch for Eniola Olutogun - [ ] Prepare biographical sketch for proposed PI - [ ] Complete NSF budget form with detailed justification - [ ] Obtain current and pending support statements from all personnel - [ ] Submit through NSF FastLane or Research.gov - [ ] Verify deadline on programme website - [ ] Confirm that Eniola can be listed as key personnel or co-PI despite not being U.S.-based - [ ] Prepare data management plan per NSF requirements - [ ] Prepare mentoring plan if required by host institution EDITOR NOTES - Eligibility risk: NSF grants typically require the PI to be at a U.S. institution. Eniola is an independent researcher in Nigeria. The proposal must be submitted through a U.S. academic host who serves as PI. Confirm with the host institution that Eniola can be listed as co-PI or senior personnel. Some NSF programs allow foreign co-PIs but this must be verified. - The EBBS programme page should be checked for specific solicitation language about international collaborators. Some NSF programs restrict funding to U.S. institutions only, though foreign collaborators can receive subawards. - The budget figures are estimates. Eniola should obtain actual quotes for HPC credits and conference travel from the host institution. - The clinical trial protocol mentions fMRI. Confirm availability of fMRI facilities in Lagos or adjust to behavioral measures only. - Eniola should verify that the provisional patent filing does not create conflict with NSF intellectual property requirements. NSF generally allows patent filings but requires disclosure.